Class Ia phosphoinositide 3-kinases (PI3Ks) mediate cell and tissue growth downstream of insulin, IGF-1 and other growth factors. Impairments in the ability of insulin to stimulate PI3K in liver and muscle results in insulin resistance and type 2 diabetes, while genetic aberrations that result in hyper-activation of PI3K in epithelial tissues results in cancers. Activating mutations in PIK3CA, the gene encoding the p110? catalytic subunit of PI3K, are among the most frequent oncogenic occurrences in epithelial cancers, and inactivating mutations in PTEN, the phosphatase that degrades the lipid product of PI3K is among the most frequent tumor suppressor mutations. Previous research funded by this grant has helped to elucidate the biochemical mechanism by which PI3Ks are activated downstream of insulin and other growth factors and to define the downstream signaling pathway that control cell growth. However, much remains to be discovered. During the next granting period we will attempt to explain some recent surprising discoveries about regulation of PI3K in disease states. The first aim is evaluate the role of the PI-3, 4-P2 4-phosphatase, INPP4B in cellular regulation and tumor formation. During the previous granting period we discovered that INPP4B acts as a tumor suppressor in breast and ovarian cancers by dephosphorylating the lipid, PI-3, 4-P2. We will characterize mice with germline deleted and floxed-INPP4B in regard to tumor formation in the context of BRCA1 and TP53 deletions. We have also observed dramatic enhancement in macrophage phagocytosis in the INPP4B knockout mice and will further characterize the role of INPP4B in macrophage biology. The second aim is to evaluate the role of the p85 subunit of Class 1a PI3Ks in BRD7 function. Recently we discovered that the regulatory subunit of PI3K (p85?) forms a tight complex with the bromo-domain protein, BRD7, a protein that mediates p53- dependent stress responses. We will further characterize the BRD7/p85 complex and determine its role in p53 and swi/snf regulation of gene expression. The final aim is to evaluate the role of novel mutations in PIK3CA and PIK3R1 in endometrial cancers. We and others have found that while breast cancer have very high rates of mutations in the catalytic and helical domains of PIK3CA (H1047R and E545K) endometrial cancers have unusually high rates of mutations in the N-terminal p85-binding domain of PIK3CA (especially R88Q), an event rarely seen in breast cancer. Endometrial cancers also have frequent mutations in the gene encoding p85? (PIK3R1). We will attempt to understand the biochemical mechanism by which these mutations contribute to endometrial cancers.